I am a guy from southern germany. I like scouting, trains and computers. Politically, I would consider myself as a democratic socialist If you wanna know more about me, have a look at: My selfhosted linkstack: https://links.strawberrycloud.org/@Straw(berry)man Or my blog: blog.strawberrycloud.org

  • 4 Posts
  • 25 Comments
Joined 3 months ago
cake
Cake day: July 29th, 2025

help-circle
  • To your first question: Google released a list of all “certified” android devices and it’s basically every phone from every halfway known brand. So yeah, you will be effected. The only devices unaffected by this would probably be no name Chinese phones (probably also Huawei, but I am not shure) and IOT devices like smart fridges. The best way to avoid this would probably be installing a custom ROM, like Graphene OS.

    To your second question, the Android System already controlls the package Installation process, do you know the “Do you want to install this APK” popup, you geht every time you want to install an app outside of the playstore? That’s controlled by the android operating systen and by extension Google. In the future, every android apk would have to have a unique “developer key” attached to it and if it isn’t verified by google, the android system can just refuse to install the apk. For that, you don’t have to go through the playstore, but you still would have to go through a verification process with Google for every app, you make. How that will be implemented in detail is not yet quite known.

    Google could have done this much earlier, it isn’t hard to implement, but you can’t make it in a way that only negatively impacts ransomware or pirated apps. And most sideloading on Android is perfectly legitimate, so the reason, why Google hasn’t done it, because there is (deservately) a big pushback from developers.













  • Nope, you can’t train a good diffusion model from scratch with just a few thousand images, that is just delusion (I am open for examples though). Adobe Firefly is a black box, so we can’t verify their claims, obviously they wouldn’t admit, if they broke copyright to train their models. We do however have strong evidence, that google, openai and stability AI used tons of images, which they had no licence to use. Also, I still doubt that all of the people, who sold on Adobe Stock either knew, what their photos are gonna be used for or explicitly wanted that or just had to accept it to be able to sell their work.

    Great counterargument to my first argument by the way 👏